Digital adoption platforms solved a real problem. They solved it with tooltips, and that worked well enough for long enough that an entire category was built around the idea. Then SAP paid $1.5 billion to acquire the best tooltip company in the world, and the category started dissolving.
Hyper is an AI onboarding agent for SaaS that does 1-on-1 screen-sharing calls with users, seeing their screen, controlling their browser, and guiding them via real-time voice. We study this market because we operate inside it. This is our read on where DAPs are headed and why.
The Accepted Wisdom: DAPs Are the Answer to Software Adoption
The story the digital adoption platform market tells goes like this.
Software is hard to use. Training doesn’t scale. You can’t sit a person next to every user. So you build a guidance layer on top of the software, tooltips and walkthroughs and checklists, and that layer tells users what to do at each step. You measure completion rates. You update the content when the product changes. Problem solved.
It’s a credible story. WalkMe was founded in 2011 and spent 13 years making it credible. By 2024, the company reported approximately $276 million in annual recurring revenue, with 42 customers paying more than $1 million per year. SAP paid $1.5 billion to acquire WalkMe in September 2024, validating the category at the highest level.
The DAP market, by analyst estimates, sits at roughly $1.2 billion today and is forecast to grow at a CAGR of 13-22% through 2033. Every report says the category is expanding.
Here is what those reports miss.
Why the Accepted Wisdom Is Wrong
Growth metrics measure the past. They count contracts signed, renewals closed, buyers who haven’t yet reconsidered their options. What they don’t measure is the structural problem underneath the category, the one that has always been there and is now being exposed by something better.
The structural problem: a tooltip tells. It does not do.
A WalkMe walkthrough says “click this button.” It does not click the button. A Whatfix tooltip says “fill in this field.” It does not fill in the field. Every DAP in existence, regardless of how sophisticated its analytics or how AI-enabled its interface, delivers guidance the user must act on alone. The software is still foreign. The user is still stuck. The tooltip is a sign pointing to a cliff you have to climb yourself.
This matters more than the market reports acknowledge. Consider the actual numbers on onboarding outcomes. Roughly 75% of new SaaS users abandon products within their first week. Onboarding completion rates across SaaS average between 40-60% for healthy products, with a median checklist completion rate of just 10.1%. Approximately 68% of users cite poor onboarding as the reason they leave a product.
DAPs have been available for over a decade. These numbers are where they are with DAPs deployed.
The category was not a failure. It was a ceiling.
The Evidence: Where DAPs Actually Break Down
There are three places where the tooltip model breaks, structurally, not incidentally.
First: walkthroughs depend on a user who is already oriented.
A tooltip assumes the user knows what they want to do and just needs help with the mechanics. It says “to create a new report, click here.” It cannot recognize when a user doesn’t know they want a report. It cannot ask “what are you trying to accomplish?” It cannot adapt when the user’s goal is one thing and the workflow they’re following is another. The tooltip is reactive. It waits to be triggered. It cannot understand context it was not pre-scripted to address.
Second: walkthroughs break when products change.
Every DAP anchors its tooltips to specific UI elements, CSS selectors or DOM structures, in the target application. When the product ships a redesign, moves a button, renames a field, or restructures a form, the walkthrough breaks. Someone has to find each broken element and fix it manually. For a company running 150 walkthroughs across 8 products, this is a permanent maintenance cycle with no end state. The product keeps shipping. The walkthroughs keep breaking. The team keeps patching.
Third: the interaction is fundamentally passive.
When a person is confused, they ask a question. A tooltip cannot answer an unanticipated question. It can only surface the answer it was built to surface. If a user asks “why do I need to do this step?” the tooltip has no response. If a user says “I already did this yesterday, why am I seeing it again?” the tooltip does not know. The information flows one direction: from the pre-scripted guidance library to the user. What the user says back, what they actually do, what they’re confused about right now, goes nowhere.
The SAP acquisition signals what this means at the category level.
When the market leader gets acquired by an incumbent enterprise software giant, one of two things is happening. Either the category is so strategically valuable that SAP paid $1.5 billion to control it. Or the category is consolidating because it has reached the limits of what its core model can do, and the most logical exit for the category leader is to become a bundled feature inside a larger platform.
The answer is probably both. But the second half of that answer is not bullish for independent DAPs.
Post-acquisition, WalkMe is being positioned primarily as the adoption layer for SAP’s own software ecosystem. That is a narrowing, not an expansion. Non-SAP customers evaluating WalkMe today are buying a tool whose product roadmap will increasingly be shaped by SAP’s priorities, not theirs.
Whatfix, the most serious remaining independent DAP vendor, responded by shipping four named AI agents. That is a telling move. The category knows what it needs to become. It is racing toward it from a standing start.
What Replaces It: AI That Does the Work Alongside the User
The constraint that made tooltips necessary was scale. You cannot hire enough people to sit with every user during their first three sessions. So you built content libraries and deployed them as overlays. Tooltips were the only thing that could approximate guidance at software scale.
That constraint no longer holds.
AI can now see a screen, understand what is on it, control a browser, hold a voice conversation in any language, and respond to what is actually happening in real time. Not next-best-action suggestions. Not a smarter tooltip. A live call with someone who can see your screen, click things for you, explain why each step matters, and answer your question when you ask it.
This is what Hyper does. One line of JavaScript to integrate. No walkthroughs to build. No content to maintain. No CSS selectors to break. When your product ships a redesign, Hyper reads the live product the same way a person would: it looks at what’s there and works with it. When a user says “wait, why am I doing this?”, Hyper answers.
The before/after contrast here is not subtle. Before: a product team spends weeks building a walkthrough library, a Customer Success team monitors completion rates, someone rebuilds the broken tours after the next release. After: the AI joins each user, handles each session, adapts to each product change, and the team looks at outcomes rather than tooltip click-through rates.
The reason DAPs existed was that no better alternative existed. That reason is gone.
What This Means for Software Teams Right Now
The DAP category is not disappearing this quarter. Enterprise contracts are sticky. WalkMe’s installed base inside SAP accounts will persist for years. Mid-market DAP vendors will continue finding buyers who haven’t yet evaluated the AI alternative.
But the trajectory is set.
Any SaaS company evaluating a DAP purchase in 2026 is buying a model that was invented to solve a constraint that technology has now removed. That is a defensible choice in specific circumstances: very large enterprises with complex cross-application workflows, strong IT governance requirements, and compliance needs that only a WalkMe-class tool can satisfy. For those buyers, the category still makes sense.
For everyone else, the question to ask is not “which DAP is best?” It is “do I want to spend $24,000-$100,000 per year building and maintaining content libraries to tell users what to do, or do I want to deploy an AI agent that does the call with them instead?”
If your goal is trial-to-paid conversion, feature adoption, or reducing time-to-value for new customers, the answer to that question is increasingly obvious.
See also: How DAP maintenance costs compound over time, Why product tours fail at activation, and our analysis of 46 tools in the onboarding and adoption space.